[EMBOSS] Bug in 'remap' program? - Checked by AntiVir DEMO version

Jean Mao maoj at helix.nih.gov
Wed Jan 17 13:30:45 UTC 2007


 Hi Guy,

There has inconsistency in the result I get. You may want to run remap with
the sequence I provide below to see the problem.

on the commend line, I run :

% remap -opt

accept all the default using the file provided below, part of the output
file look like this (which I don't see when not using -opt flag, why?) :

===============================================================
# Enzymes that cut  Frequency   Isoschizomers
      AluI          1   MltI
      ApaI          1   PpeI
      AsuI          2
AspS9I,AvcI,Bac36I,Bal228I,BavAII,BavBII,Bce22I,BshKI,BsiZI,Bsp1894I,BspBII,
BspF4I,Bsu54I,CcuI,Cfr13I,MaeK81
II,Nsp7121I,NspIV,Pde12I,PspPI,Sau96I
      BfiI          1   BmrI,BmuI
  BmgT120I          2   
     BseSI          1   Bme1580I
     BsiYI          1   BflI,Bsc107I,Bsc4I,BseLI,AfiI,BslI,Bst22I
   Bsp120I          1   PspOMI
      BsrI          1   BseNI,Bse1I,BsrSI,Bst11I,Tsp1I
     Csp6I          1   CviQI,CviRII
     CviJI          2   CviKI,CviKI-1
     CviRI          1   HpyCH4V,HpyF44III
     DraII          1   EcoO109I
      FmuI          2   
    HaeIII          1
BecAII,Bim19II,Bme361I,BseQI,BshFI,BshI,BsnI,Bsp211I,BspANI,BspBRI,BspKI,Bsp
RI,BsuRI,BteI,CltI,DsaII,EsaBC4I
,FnuDI,BanAI,MchAII,MfoAI,NgoPII,NspLKI,PalI,Pde133I,PflKI,PhoI,PlaI,Pru2I,S
bvI,SfaI,SuaI
    HgiJII          1
BpuI,Bsp519I,Bsu1854I,BvuI,Eco24I,Eco75KI,EcoT38I,FriOI,BanII,KoxII,PaeHI,Sa
cNI
     Hpy8I          1   HpyBII
   Kaz48kI          1   PssI
     NlaIV          1   BmiI,BscBI,BspLI,AspNI,PspN4I
      PabI          1   
      RsaI          1   HpyBI,PlaAII,AfaI
      SduI          1   BmyI,BsoCI,Bsp1286I,BspLS2I,MhlI,NspII,AocII
     TspRI          1   
      UnbI          2   
==========================================================================


As you can see, 6 enzymes show NO isoschizomers. I assume all of them have
commercial supplier(s) since I accept the default setting. However, using
'redata' program in EMBOSS on these 6 enzymers, some of them DO have
isoschizomers but the field was left blank. In addition, some of them has NO
suppliers listed which is not suppose to appear when I use the default
settings, isn't it?

Thank you in advance.


================= Sequence I Used
=============================================


!!NA_SEQUENCE 1.0
LOCUS       A00006                    26 bp    DNA     linear   PAT
10-FEB-1993
DEFINITION  Artificial oligonucleotide sequence (Fra 3), sequence 5 from
patent
            application EP0238993.
ACCESSION   A00006
VERSION     A00006.1  GI:57973
KEYWORDS    .
SOURCE      synthetic construct
  ORGANISM  synthetic construct
            other sequences; artificial sequences.
REFERENCE   1  (bases 1 to 26)
  AUTHORS   Auerswald,E.A., Schroeder,W., Schnabel,E., Bruns,W.,
Reinhardt,G.
            and Kotick,M.
  TITLE     Aprotinin homologues produced by genetic engineering
  JOURNAL   Patent: EP 0238993-A 5 30-SEP-1987;
            BAYER AG
FEATURES             Location/Qualifiers
     source          1. .26
                     /organism="synthetic construct"
                     /mol_type="unassigned DNA"
                     /db_xref="taxon:32630"
ORIGIN

  A00006  Length: 26  January 11, 2007 09:44  Type: N  Check: 4746  ..

       1  CGCCGTACAC TGGGCCCTGC AAAGCT


-----Original Message-----
From: Guy Bottu [mailto:gbottu at ben.vub.ac.be] 
Sent: 2007年1月17日 3:59
To: Jean Mao
Cc: emboss at lists.open-bio.org
Subject: Re: [EMBOSS] Bug in 'remap' program? - Checked by AntiVir DEMO
version 

	Dear Jean,

The program remap by default only outputs one representative member (the
prototype) of a series of isoschizomers and it only considers enzymes that
have a commercial provider. If you want to see all enzymes you must run
remap with parameters -nolimit -nocommercial.

	Regards,
	Guy Bottu,
	Belgian EMBnet Node



More information about the EMBOSS mailing list